New book about Kelly criterion

[GameTheory]

Fortune's Formula : The Untold Story of the Scientific Betting System That Beat the Casinos and Wall Street

by

William Poundstone

I haven't read it, but knowing Poundstone this should be a good read for those interested in such things:

http://www.amazon.com/exec/obidos/tg/detail/-/0809046377/qid=1126579789/sr=8-1/ref=pd_bbs_1/002-0470502-4055218?v=glance&s=books&n=507846

[GameTheory]

While we're at it, this one looks interesting too. Again, haven't read it:

http://www.amazon.com/exec/obidos/tg/detail/-/0471602450/qid=1126580076/sr=12-1/002-0470502-4055218?v=glance&s=books

[Dave Schwartz]

GT,

Thank you for the suggestions.

I have ordered both.

Dave Schwartz

[CapperLou]

Thanks for heads up on these. Just ordered both too.

CapperLou

[Kreed]

I'm also interested in both books but I doubt if any good read will beat any
casino game --- and will NEVER lead to better investments on Wall St.
My practical brain goes into ALERT when I hear these hypes. And just my
personal PS: No One has ever seen Thorpe's Tax Returns --- so the academic
turned gambler sounds so inviting, but where's any proof. OTOH, playing his
Basic Strategy by counting 5's & 6's etc etc just passes time before disaster.

[Dave Schwartz]

My Review of Mean Markets & Lizard Brains

The first few pages address "Lizard Brains," an interesting concept that I had hoped would have some connection to what we do. (It does, but fairly simple.)

Basically, Lizard Brains refers to our instinct that tells us when something is a good or bad bet. The author sights studies which show that when asked if a current investment makes the owner feel confident or unsure, the confident investments are more likely to go down and the unsure investments are more likely to go up.

Mean Markets refers to the tendency of the stock market to seem to be aimed precisely at screwing up the individual investor.

Both of these items are something that most horseplayer can identify with.


The explanation of the inter-dependencies between stocks, currencies, bonds and deficits was very educational as well.



Unfortunately, the rest of the book is full of contradictory platitudes and half-truths that I just cannot learn anything from.

For example, the author spends a great deal of time showing that people who buy-and-hold will always do just fine in the stock market, no matter whether they bought at a top or a bottom (providing they are somewhat diversified). Then, late in the book he says, "Don't purchase on the way down."

Well, if it is okay to buy at the bottom but not okay to buy while a stock is on the way down, then that must mean you must somehow tell when the market is down but will not go down further.


All in all, I found the book informative and educational, but do not expect anything huge in the way of improvement over what you are doing now, whether in the markets or horse racing.


I would rate this book C+.


Dave Schwartz

[Overlay]

Dave:

Will you still be reading and reviewing Fortune's Formula also?

[Dave Schwartz]

I just began reading it today.

[Kreed]

Let me raise my kilt to the individual investor. He gets stiffed at every turn.
There are only very few such players who can turn a ROI > 8%. Its a fact,
that firms follow closely. It's no poor reflection on the investor, but rather
the game he's playing is tilted. (Just my opinion.)

[fastCow]

One of the more intriguing books about analyzing the market is by Benoit Mandelbrot, "The (Mis)Behavior of Markets A Fractal view of Risk, Ruin, and Reward."

Among his points is that Gaussian (the world is a neat grid) techniques do not accurately portray market turbulence. An example of this is the Black-Scholes model (discussed in Fortune's Formula) for option pricing, which assumes constant volatility (the world does not change).

A mathematical theorist, he does not claim to have a method for beating the market. But he does have a convincing argument that there are models based on fractal mathematics and power laws give a truer picture of market behavior. He has some valuable ideas regarding trading time, fat tails in bell curves, long memories, and so on.

I think some of this can be related to handicapping. For instance, his claim that the bell curve model is inaccurate for turbulent markets makes sense for handicappers. In a bell curve most of what you are analyzing has to be in the middle or close to it. The activities at the ends (tails) are supposed to be neglible. But he proves that is not the case in markets, hence "fat tails."

The few (count on 3 fingers) people that I know who have managed to make over $60k a year at the track tell me that their profits occur in a small percentage of their wagers. The rest of the time they are basically treading water. I have found that it is not that difficult to find a computer model of a profitable wagering strategy. The rub is that, typically, the profit occurs in a small percentage of the wagers.`In other words, the profit comes in the fat tails. And it is very difficult to predict if the complex factors that contributed to those fat tails will repeat themselves.

So unless wagering strategy addresses this distribution, it will be difficult to predict the capital (bankroll) required to sustain you while you wait for those profitable situations to occur. Maybe that is why they call it gambling.

[jfdinneen]

fastcow,

I agree with your positive appraisal of Mandelbrot's book. In similar vein, I would also recommend Paulos's "A Mathematician Plays the Market" (e.g., Average Riches, Likely Poverty pp 95-99).

If you wish to explore a mechanism for combining multiple losing strategies into a winning equivalent, I would encourage you to look at Parrondo's Paradox (http://www.cut-the-knot.org/ctk/Parrondo.shtml) as a possible way forward. Personally, I am successfully exploiting a combination of laying pretenders on Betfair markets with backing contenders in parimutuel exotics markets.

Best wishes,

John

[Dave Schwartz]

Dang, you guys are ruining my life with all these great book ideas. I haven't read this much in 20 years.

Okay, back to Amazon.

[GameTheory]

Just wait for the upcoming thread, "GameTheory's Complete Interdisciplinary Library for the Handicapper and Bettor" where you will be forced to read (for example):

Against The Gods (http://www.amazon.com/exec/obidos/tg/detail/-/0471295639/qid=1128446510/sr=8-1/ref=pd_bbs_1/104-8063994-1556711?v=glance&s=books&n=507846) by Peter L. Berstein

Entanglement (http://www.amazon.com/exec/obidos/tg/detail/-/0452284570/qid=1128447061/sr=2-1/ref=pd_bbs_b_2_1/104-8063994-1556711?v=glance&s=books) by Amir Aczel (beginners will be referred also to Aczel's probability theory primer Chance (http://www.amazon.com/exec/obidos/tg/detail/-/1843440229/qid=1128447136/sr=1-3/ref=sr_1_3/104-8063994-1556711?v=glance&s=books) , although there are better ones out there)

and the sprawling The Education of a Speculator (http://www.amazon.com/exec/obidos/tg/detail/-/0471249483/qid=1128447346/sr=2-1/ref=pd_bbs_b_2_1/104-8063994-1556711?v=glance&s=books) by Victor Niederhoffer

[jfdinneen]

GameTheory,

I concur with your recommendation of both "Against The Gods" and "Education of a Speculator". May I add Fooled By Randomness (http://www.amazon.com/gp/product/0812975219/104-7336663-0863968?v=glance&n=283155&s=books&v=glance) by Taleb who claims to have been on the other side of many of Niederhoffer's biggest transactions and who has an alternate description of "fat tails" - namely, "Black Swans"!

Best wishes,

John

[GameTheory]

Yeah, Fooled by Randomness is a must-read for horseplayers...

[Dave Schwartz]

GT,

Stop! Stop!

(I have read several of those already, btw.)


Dave

[sealord]

I'm just finishing up 'Fortune's Formula' and man, what a phenomenal read. It reads like a novel, but is more a collection of important works discovered by great, great men. The applicability of the info discussed is pertinent, largely due to massive discussions of the Kelly criterion, but I learned a bit about Wall Street too. It was the first of these type books I've read, and will rabidly consume the others you fine members mentioned in this thread. Thanks for turning me on the this book.

[andicap]

Given Dave S.'s take on Kelly I'm eager to hear his opinion on the Poundstone book.

[Dave Schwartz]

I have only read a couple of chapters thus far, but I can tell you it is very enjoyable so far.


Dave

[highnote]

Here is a link to a review of Poundstone's book

> > > http://www.americanscientist.org/template/BookReviewTypeDetail/assetid/47321
> > >
> > > Bettor Math
> > > Elwyn Berlekamp
> > > Fortune's Formula: The Untold Story of the Scientific Betting
> > > System that Beat the Casinos and Wall Street. William Poundstone.
> > > x
> > >
> > > + 367 pp. Hill and Wang, 2005. $27.
> > >
> > > Every investor must decide how to partition her portfolio among
> > > many possible investments. Plausible strategies range from
> > > "diversify" to "focus."
> > >
> > > In a paper published in 1956, John L. Kelly of Bell Labs
> > > formulated the asset-allocation problem in terms of an idealized
> > > model for which he derived some quantitative results. He used
> > > colorful racetrack terminology reminiscent of the classic Damon
> > > Runyon movie Guys and Dolls: Suppose that one goes to the
> > > racetrack with an available bankroll,B. Suppose further that one
> > > knows for each horse the correct probability that it will win the
> > > next race. Suppose further that the betting odds are at least
> > > slightly inconsistent with this information. And finally, suppose
> > > that each race is merely one of a very long sequence of betting
> > > opportunities. Kelly found criteria for deciding how much one
> > > should then bet on each horse in each race.
> > >

The review continues...

> > > Ed Thorp analyzed the game of blackjack far more deeply than
> > > anyone had ever done before, and he devised card-counting schemes
> > > to gain an edge, especially toward the end of a deck that is not
> > > reshuffled after every deal. He wrote a bestseller, Beat the
> > > Dealer, on how to win at blackjack. Earlier in his career, when he
> > > was a mathematics instructor at MIT, he met Claude Shannon, and he
> > > brought Claude and Betty Shannon with him as partners on one of
> > > his early weekend forays to Las Vegas. Later, he discovered and
> > > exploited a number of pricing anomalies in the securities markets
> > > and made a significant fortune. Thorp's first hedge fund,
> > > Princeton-Newport, achieved an annualized net return of 15.1
> > > percent over 19 years, and in May 1998, Thorp reported that his
> > > own investments had an annualized 20 percent return over 28.5
> > > years.

[Buckeye]

If the bankroll can survive long enough, good things can happen.

The trick is looking back and actually having been there!

Bets must be very small in relation to the bankroll, it's that simple.

[arkansasman]

I read the book a couple of weeks ago and thought it was excellent. But, in disagreement with some, I am a believer of Kelly and use it in my model. I thought so much of the book, I recommended it to one of the world's greatest horse players.

[highnote]

I thought so much of the book, I recommended it to one of the world's greatest horse players.

I was wondering how Dave Schwartz found out about it so quickly! ;)

[highnote]

RE Neiderhoffer's -- Education of a Speculator

After reading a chapter or two I got the feeling this guy was going to crash and burn.

He did.

He is obviously very bright and I pay attention to what he says when I see him quoted. But I didn't think his book was that good. I didn't get any good ideas for trading from his book. But I don't consider myself a speculator, so maybe that's why his book didn't appeal to me.

I'd rather do arbitrage and have a guaranteed profit than to bet with a chance of losing. The caveat to that, though, is if I know a long string of investments where some win and some lose will outperform an arbitrage only approach.

However, since my net worth pales in comparison to Ed Thorp's, you can discount my investment advice by a factor of about 100. Probably more -- since his wealth is probably in liquid securities and mine is mostly in my house.

[toetoe]

Niederhoffer is at least a free thinker. Too few around these days. As to crashing, doesn't everybody, at some point? He's not eating cat food for the rest of his life, is he?

[highnote]

Niederhoffer is at least a free thinker. Too few around these days. As to crashing, doesn't everybody, at some point? He's not eating cat food for the rest of his life, is he?

I think he said in his book that he put aside enough money so that if he lost all his bankroll he'd still have enough left over to be comfortable. He told his wife to never, ever, ever let him touch that money to use for gambling.

That was smart.

Obviously he is very sharp. Donald Trump is very sharp, too, and he has had bankruptcies. Guys like them have a knack for making comebacks. I'm sure Neiderhoffer is doing OK.

[Kreed]

Donald Trump? Is he (could he be) anyone's business idol? He's a loser
in every way, and boring too. His son, Donald Jr (LOL, see what I mean)
is very different, but stuck with his biology. When will fathers grow up & stop
trying to mold their sons? And Wall St, does it have any lessons to teach,
for the individual investor? ummm, its the rare bird that gets the worm there.

[highnote]

Donald Trump? Is he (could he be) anyone's business idol? He's a loser
in every way, and boring too. His son, Donald Jr (LOL, see what I mean)
is very different, but stuck with his biology. When will fathers grow up & stop
trying to mold their sons? And Wall St, does it have any lessons to teach,
for the individual investor? ummm, its the rare bird that gets the worm there.


Kreed,

I don't know about you, but I kind of like "The Donald".

You say he's a loser in every way and boring, too. That reminds me of my old signature line from H.G. Wells... "Righteous indignation can be traced to envy." Could it be you wish you had a little bit of Trump-like ability?

So, it sounds like when you say "it's the rare bird that gets the worm on wall street", you are implying that individual investors can't or don't very often make money investing in stocks. If that's what you mean, then you are way wrong. But, from my experience, you usually are. So as usual, I will consider the source.

John

[Kreed]

SwetYeJohn, Don't you work? No time for that with all your indignations?
Try 2 read better too, As in asking me, after I scorned Donald Trump, "I don't
know about you (Derek), but I kinda like the Donald." --- Of course, I'll be
at work, protecting the USA economy & this noble land's cherished freedoms
while you're spinning a reply, with all that time on your hands.

[highnote]

SwetYeJohn, Don't you work?

What makes you think I don't work? Because I post on PA?

No time for that with all your indignations?

That's an interesting comment. Do you know the definition of indignant? In case you don't, it is "Anger aroused by something unjust, mean, or unworthy".

So if I'm indignant you must have done something unjust, mean or unworthy.

So again I say, I just consider the source.

Try 2 read better too, As in asking me, after I scorned Donald Trump, "I don't know about you (Derek), but I kinda like the Donald."

I have no idea what you're talking about. What or who is Derek?



--- Of course, I'll be
at work, protecting the USA economy & this noble land's cherished freedoms

Where's that damn vomit icon, PA?

[PaceAdvantage]

I'm fed up with the horse's asses.....


You mentioned vomit?? I agree...time to purge....

[Macdiarmadillo]

Tons of good stories, a very good read. Not a whole heckuva lot about horse racing application, though. Nor is there a lot of detail on approaches; somebody made a lot of money and somebody else didn’t is about what I read. For example. did everyone mentioned use “straight” Kelly and never partial? The book does give source citations; maybe what I'm looking for is there.

It did briefly make the point about large bets based on Kelly then further affecting the odds. I think this is very much the case at most tracks where WPS is an ever smaller pool compared to the exotics. You also have the situation where a big chunk of the pools drops in after you can bet. Anyone having a model where they can predict what the actual pools will be based on totals X minutes before post, please speak up.

So we have the situation where Hong Kong has giant pools and had groups that are were using Kelly. If they were running concurrently (as it sounds like), then you have Group A plunking in their bets first, then Group B analyzing the pools after that point would probably determine “no bet” assuming Group A’s bets being large enough to affect the pool. You’d probably play that way assuming a big enough bankroll. Then assuming A got their bets in front of B consistently, then wouldn’t B try to bet earlier than A, and back and forth until either one was looking at pools so early that the info was unpredictive. Or are HK pools so large that betting from multiple groups would not affect payouts?

[GameTheory]

Or are HK pools so large that betting from multiple groups would not affect payouts?They probably didn't come up with the exact same bets, and yes, the HK pools are astronomical.

[Hosshead]

Big Pools- Big Bets.
Even the largest (HK) pools can be changed by big bets.
The (t.v.) show I saw, had the HK Handicapping syndicate "Honcho", (Austrailian) but living in Hong Kong, betting Millions / per DAY !

[formula_2002]

Tons of good stories, a very good read. Not a whole heckuva lot about horse racing application, though. Nor is there a lot of detail on approaches; somebody made a lot of money and somebody else didn’t is about what I read. For example. did everyone mentioned use “straight” Kelly and never partial? The book does give source citations; maybe what I'm looking for is there.

It did briefly make the point about large bets based on Kelly then further affecting the odds. I think this is very much the case at most tracks where WPS is an ever smaller pool compared to the exotics. You also have the situation where a big chunk of the pools drops in after you can bet. Anyone having a model where they can predict what the actual pools will be based on totals X minutes before post, please speak up.

So we have the situation where Hong Kong has giant pools and had groups that are were using Kelly. If they were running concurrently (as it sounds like), then you have Group A plunking in their bets first, then Group B analyzing the pools after that point would probably determine “no bet” assuming Group A’s bets being large enough to affect the pool. You’d probably play that way assuming a big enough bankroll. Then assuming A got their bets in front of B consistently, then wouldn’t B try to bet earlier than A, and back and forth until either one was looking at pools so early that the info was unpredictive. Or are HK pools so large that betting from multiple groups would not affect payouts?


I was just about to post this on my "..with out figures post"

What can I hope to prove, when, in W.Poundstone’s recent book “Fortune’s Formula”, Thorp states;

“I estimate that ….I have made $80 billion worth of purchases and sales (‘action’, in casino language) for my investors. This breaks down to something like one and a quarter million individual ‘bets’ averaging $65,000 each …. . Over all, it would seem to be a moderately ‘long run’ with a high probability that the excess performance is more than chance”


I marveled at the "it would seem".

[highnote]

I marveled at the "it would seem".

Infinity is a long time. :D

[formula_2002]

"Things that come with little help from luck are more resistant to randomness" -- Nassim Taleb

I read his stuff also.
Did I miss his proof?

[highnote]

"Things that come with little help from luck are more resistant to randomness" -- Nassim Taleb

I read his stuff also.
Did I miss his proof?


I don't know how to answer that. :confused:

[facorsig]

I bought Fortune's Formula based on recommendation in this thread. I wish I had the time to read in a limited number of sittings, but instead I read small segments every day. I thought the book was fascinating. Starts with some of the origins of wire services (transmitting race results for wagering where the results were not yet known), their links to organized crime (and the Daily Racing Form). and goes into the mathematical questions about maximizing wealth and utility. Towards the end, the book spends an increasing amount of time on the stock market and characters such as Rudy Giuliani, Michael Milken and Ivan Boesky. Ed Thorp is a constant throughout the book.

The writer's style is outstanding. The book is composed into a series of short chapters on a concept with a surrounding, typically non-fictional, vignette.

In addition to the Kelly criterion, there was a tremendous content of reviews of money management concepts and their (mis)application. While the content is not 100% horse racing related, I think it is relevant, particularly for money management.

[DJofSD]

I'm only part way into the book but so far it's an enjoyable read. The style is some what like that of Simon Winchester's but at times a little strained.